Reactions of copper(II) chloride in solution: Facile formation of tetranuclear copper clusters and other complexes that are relevant in catalytic redox processes

Article abstract of DOI:10.1002/chem.201203848

Mixing CuCl₂·2 H₂O with benzylamine in alcoholic solutions led to an extremely colorful chemistry caused by the formation of a large number of different complexes. Many of these different species could be structurally characterized.

Full text of DOI:10.1002/chem.201203848

DOI: 10.1002/chem.201203848
Reactions of Copper(II) Chloride in Solution: Facile Formation of
Tetranuclear Copper Clusters and Other Complexes That Are Relevant in
Catalytic Redox Processes
Sabine Lçw,[a] Jonathan Becker,[a] Christian Wꢀrtele,[a] Andreas Miska,[a]
Christian Kleeberg,[b] Ulrich Behrens,[c] Olaf Walter,[d] and Siegfried Schindler*[a]
Abstract: Mixing CuCl2·2H2O with
benzylamine in alcoholic solutions led
to an extremely colorful chemistry
caused by the formation of a large
number of different complexes. Many
of these different species could be
structurally characterized. These in-
clude relatively simple compounds
such as [Cu(L1)4Cl2] (L1 =benzylamine)
proved to be highly reactive in a series
of oxidation reactions of organic sub-
strates by using air or peroxides as oxi-
dants. Furthermore, it was possible to
isolate and structurally characterize
tion, it was possible to solve the molec-
ular structures of the compounds
Cu4OCl
6ACHTUNGTNERNUG(MeOH)4, [CuACHTUNTGERN(NUGN MeOH)2Cl2],
[Cu(aniline)2Cl2], and an organic side
AHCTUNGTRENNUNG
product (HC13H19NOCl). In fact all de-
termined structures are of a known
type but the chemical relation between
these compounds could be explained
for the first time. The paper describes
these different compounds and their
chemical equilibria. Some of these
complexes seem to be relevant in cata-
lytic oxidation reactions and their reac-
tivity is discussed in more detail.
([Cu(L1)Cl]3 and [Cu
ACHTUNGTRENUNG(benz2mpa)2]CuCl2
(benz2mpa=benzyl-(2-benzylimino-1-
methyl-propylidene)-amine), two cop-
per(I) complexes that formed in solu-
tion, demonstrating the high redox ac-
tivity of the cluster systems. In addi-
and (HL1)
2ACHTUNGTRENNUNG[CuCl4]. Most interestingly
is the easy formation of two cluster
complexes, one based on two cluster
units Cu4OCl6(L1)4 connected through
one [Cu(L1)2Cl2] complex and one
Keywords: cluster compounds
·
copper · homogeneous catalysis ·
hydroxylation · oxidation
based on
a
cubane-type cluster
([Cu4O4](C11H14)4Cl4). Both clusters
ACHTUNGTRENNUNG
Introduction
Copper complexes that form clusters of the type Cu4OX6L4
(X=halogen, L=ligand or X) or Cu4O4L4X4 (cubane-type
structure) are already known since the 1960s.[1,2] Both basic
structural units are shown in Figure 1. Such multicopper
[a] S. Lçw, J. Becker, C. Wꢀrtele, A. Miska, Prof. Dr. S. Schindler
Justus-Liebig-Universitꢁt Gießen
Institut fꢀr Anorganische und Analytische Chemie
Heinrich-Buff-Ring 58, 35392 Gießen (Germany)
Fax : (+49)641-99-34149
E-mail: siegfried.schindler@chemie.uni-giessen.de
Figure 1. Schematic presentation of Cu4OX6L4 (left) and the cubane unit
Cu4O4L4X4 (right) (e.g., X=chloride).
[b] Dr. C. Kleeberg
Technische Universitꢁt Braunschweig
Institut fꢀr Anorganische und Analytische Chemie
Hagenring 30, 38106 Braunschweig (Germany)
compounds have been investigated especially with regard to
their magnetic properties.[1a,h,2b,d,e,3] However, the synthesis
of such compounds in some cases can be quite a challenge.
Furthermore, several compounds of the formula Cu4OX6L4
were sensitive towards moisture and/or oxygen and there-
fore difficult to handle.[1c,e,f]
A lack of applications and the described synthetic difficul-
ties led to a strong decrease of interest in these compounds
during the last decades. Finally, in 2002 catalytic properties
of the cubane element in redox reactions were recognized.[2c]
[c] Prof. Dr. U. Behrens
Universitꢁt Hamburg
Institut fꢀr Anorganische und Angewandte Chemie
Martin-Luther-King-Platz 6, 20146 Hamburg (Germany)
[d] Dr. O. Walter
European Commission Joint Research Centre
Institute for Transuranium Elements
Hermann-von-Helmholtz-Platz 1
76344 Eggenstein-Leopoldshafen (Germany)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201203848.
5342
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5342 – 5351

FULL PAPER
Since then multicopper com-
pounds have been used as cata-
lysts for oxidation reactions
À
such as hydroxylation of C H
bonds and oxidation of cate-
chol.[2d–f,4] However, so far there
is no detailed understanding of
the reactivity of these com-
pounds. This is quite unfortu-
nate because multicopper sys-
tems have the advantage to
provide/accept more than one
electron in a redox reaction, a
concept that is well known in
nature from the active site in
a
large number of metallo-
ACHTUNGTRENNUNG
enzymes.[5]
Here, we describe the facile
preparation of both cluster
types in solution from
CuCl2·2H2O and benzylamine.
Especially cluster I seems to be
quite useful in catalytic oxida-
tion reactions. During our in-
vestigations on the reactivity of
cluster I we observed the for-
Figure 2. Molecular structure of cluster I (hydrogen atoms and crystallized solvent molecules are omitted for
clarity). Ellipsoids are drawn at the 50% probability level.
mation of a large number of copper complexes strongly de-
pending on the reaction conditions. Most of these complexes
could be structurally characterized and the chemical rela-
tionship between these compounds could be explained.
is not sensitive towards air, which is mirrored in the facile
synthesis of cluster I. The molecular structure is quite inter-
esting because it formed a cluster dimer that is “connected”
1
À
through a [Cu(L )2Cl2] unit (the Cu5 Cl4 distance is 3.06 ꢃ
À
compared to a Cu1 Cl1 bond lengths of 2.44 ꢃ in the clus-
ter unit and 2.32 ꢃ (Cu5 Cl7) in the complex). Interestingly,
À
Results and Discussion
so far we could not detect formation of Cu4OCl6(L1)4 alone,
only cluster I formed in all our samples. The central oxide
ion in these complex units is tetrahedrally coordinated to
four copper(II) ions as described above. The oxygen atom
derives from water present in solution.[1a,f,j]
Due to our interest in copper complexes and their reactivity
towards dioxygen[6] we had started to investigate the reac-
tion of copper(II) salts with different amines. When reacting
CuCl2·2H2O with benzylamine (L1) we expected to obtain
[Cu(L1)2Cl2], a simple copper amine complex that had been
structurally described previously.[7] However, when charac-
terizing our product we observed that by chance a polynu-
clear copper complex had formed similar to previous inves-
tigations on different systems.[1f,h] Although the expected
Especially interesting was the observation that depending
on the concentrations and the solvent solutions with a huge
color range were obtained. Furthermore, in most cases color
changes occurred while standing. If the samples were not
stirred, strong color gradients within particular samples
were observed as shown in Figure 4. Especially this finding
forced us to further investigate the chemistry of these com-
plexes, however, it also clearly demonstrated the complexity
of the system.
To the best of our knowledge, investigations on the reac-
tivity of clusters of the type Cu4OX6L4 have not been report-
ed so far. Furthermore, we are only aware of one important
report by Fenton and co-workers where similar color
changes due to cluster formation were reported for cop-
[CuL1 Cl2] still formed, it co-crystallized with two
2
Cu4OCl6(L1)4 cluster units (the outcome of the reaction is
strongly dependent on the reaction conditions, see below).
The molecular structure of what we call “cluster I” is shown
in Figure 2. Crystallographic data, bond lengths, and angles
of cluster I·2CH3OH are reported in the Supporting Infor-
mation. When reacting CuBr2 instead of CuCl2·2H2O with
benzylamine the same structural unit formed, the
[Cu(L1)2Br2] complex co-crystallized with two Cu4OBr6(L1)4
cluster units (see the Supporting Information).
AHCTUNGTREGpNNUN er(II) complexes with the ligand teed (teed=N,N,N’N’-tet-
raethylethylenediamine).[8] The authors already were quite
successful in understanding the reactivity of this complex
system, however, here the situation is even more complicat-
ed due to an additional radical reaction of teed with cop-
per(II) ions leading to ligand decomposition.
Cluster I is formed without any problems in large quanti-
ties as an olive green powder (Figure 3) that deposited im-
mediately after the reaction is completed (a few seconds).
In contrast to other clusters of the type Cu4OX6L4, cluster I
Chem. Eur. J. 2013, 19, 5342 – 5351
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5343

S. Schindler et al.
(Figure 3), was formed that
turned out to be (HL1)
CHTUNGTRENNUNG[CuCl4].
2A
The molecular structure of
compound is shown in
3
Figure 6 (crystallographic data
are reported in the Supporting
Information). Here, benzyl-
AHCTUNGERTGaNNUN mine is protonated with
[CuCl4]2À as an anion. Complex
salts of this type are well
known in literature.[9]
Together with water cluster I
reacted to a turquoise powder
(compound 6, Figure 3) that so
far could not be characterized.
Very surprising was our ob-
servation that leaving suspen-
sions/solutions of cluster I in
methanol (as well as in other
solvents) caused a color change
from green to red and a red
brown powder (compound 4)
Figure 3. Overview of the color range of cluster I and related compounds (compounds that have been structur-
ally characterized are encircled).
Figure 5. Molecular structure of [Cu(L1)4Cl2] (2), hydrogen atoms omitted
for clarity. Ellipsoids are drawn at the 50% probability level.
Figure 4. Detailed view of a section in a flask containing a mixture of
benzylamine (0.2 molLÀ1), CuCl2·2H2O (0.1 molLÀ1) in denatured etha-
nol (containing 2-butanone); the mixture was not stirred for a day so that
the whole color gradient became visible. The different species could be
isolated and crystallized as described in the text.
could be isolated. Enforcing the reaction by bubbling air
through the solution allowed us to isolate an orange brown
solid (compound 5, Figure 3). Unfortunately, so far despite
much effort neither compound 4 nor compound 5 could be
identified when obtained from solutions in methanol. In this
context it is very important to distinguish between meth-
Adding benzylamine in slight excess to cluster I first led
to a green material (compound 1, Figure 3) that could be
structurally
characterized.
As
primarily
expected
ACHTUNGTRENNUNG
[Cu(L1)2Cl2] (1) is formed that has been described previous-
ly.[7] If benzylamine was added in a large excess compound 2
formed, a blue complex (Figure 3) that could be structurally
characterized and turned out to be octahedral [Cu(L1)4Cl2].
Its molecular structure is shown in Figure 5 (crystallographic
data are reported in the Supporting Information).
AHCTUNGTRENNUNG
usage as drinking alcohol (declared in Germany as dena-
tured ethanol).
Most exiting however, was the finding that a suspension
of cluster I in denatured ethanol caused the formation of a
second different cluster (“cluster II”) of the cubane type
(Figure 1). The dark-green-colored crystals that had formed
In contrast, if the amount of CuCl2·2H2O was increased a
very pretty golden compound, that is, compound
3
5344
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5342 – 5351

Reactions of Copper(II) Chloride in Solution
FULL PAPER
time that such a cluster could be obtained in a self-assembly
reaction caused by a selective oxidation reaction of an in
situ formed ligand. Here, it is very important to point out
that although hydroxylation reactions are common during
the reaction of copper(I) complexes with dioxygen, to the
best of our knowledge such a reaction so far has never been
observed starting with a copper(II) compound. This indi-
cates that it is most likely that redox reactions cause forma-
tion of copper(I) complexes that afterwards will lead to
ligand hydroxylation in the presence of dioxygen.
Support for the formation of copper(I) complexes was ob-
tained when cluster I was kept overnight in denatured etha-
nol excluding air. Again, the solution turned to a red color,
however this time a colorless material crystallized in good
yields (compound 7, Figure 3) instead of formation of clus-
ter II. These colorless crystals are very sensitive towards air
and moisture and turn to a green color under air. Despite
the fact that it was very difficult to obtain suitable crystals
for structural analysis (due to crystallization in a layered
structure), we were able to determine the molecular struc-
ture of this complex (Figure 8, crystallographic data are re-
ported in the Supporting Information).
Figure 6. Molecular structure of (HL1)
at the 50% probability level.
2ACHTUNGERTN[NUNG CuCl4] (3). Ellipsoids are drawn
The structure shows that [Cu(L1)Cl]3,
a trimer, has
formed. It is obvious that this trimer is only a small piece of
Figure 8. Molecular structure of compound 7. Ellipsoids are drawn at the
50% probability level.
Figure 7. Molecular structure of cluster II ([Cu4O4]ACHTNURTGNEUG(N C11H14)4Cl4), hydro-
gen atoms omitted for clarity. Ellipsoids are drawn at the 50% probabili-
ty level.
(Figure 3) could be structurally characterized and the molec-
ular structure of cluster II is shown in Figure 7 (crystallo-
graphic data are reported in the Supporting Information).
The new ligand is obviously formed by condensation be-
tween benzylamine and 2-butanone and selective hydroxyla-
tion of this compound by cluster I according to Equa-
tion (1):
a long linear polymeric chain in the crystals, explaining the
layered structure. It is important to point out that here weak
unsupported CuI···CuI interactions are observed that are
documented with distances of 2.88 and 2.93 ꢃ between the
copper ions. Structures of this type are interesting with
regard to the CuI···CuI interactions that are not enforced by
the ligand environment.[10] The reason for this behavior is
not quite clear and has been discussed previously in litera-
ture.[10c] Unsupported CuI···CuI interactions have been re-
ported with distances in the
Again, cluster II can be easily prepared and is obtained
reproducible in acceptable yields. This seems to be the first
range of 2.65–3.02 ꢃ in compar-
ison to the sum of the van der
Waals radii (2.80 ꢃ).[10d]
Efforts to obtain this cop-
per(I) species by reducing clus-
Chem. Eur. J. 2013, 19, 5342 – 5351
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5345

S. Schindler et al.
ter I with NaBH4 in methanol were not successful. First, a
black powder was obtained, then after prolonged stirring
The ligand of 4a is probably formed by oxidation of the
ligand of cluster II and consecutive condensation with ben-
zylamine according to Equation (2):
(10–20 min)
a
turquoise/bluish compound formed that
afterwards
turned green
(Figure 4). This cycle could be
repeated several times, howev-
er, phases to reach the different
colors became longer. Again, so
far these compounds could not
be characterized.
Further evidence for the
presence of a copper(I) species
in solution was obtained when red crystals could be isolated
together with compound 7 from a red-colored ethanolic (de-
natured) solution of cluster I. Especially here it was impor-
tant to find the right conditions and timing to isolate these
crystals from the red solution. It took a lot of time and a
large number of attempts to finally crystallize this species.
The amount of dioxygen available has a huge influence on
the outcome of the reaction (species and hence, color). If no
or only a small amount of dioxygen was available the solu-
tion turned to a red color and colorless crystals formed
(compound 7). If dioxygen is present, for example, due to
air entering the open flask, the solution again turned to a
red color, however this time dark green crystals formed
(cluster II). In both cases it was possible only after many at-
tempts to obtain a few red crystals co-crystallized with color-
less crystals (compound 7). Usually the red compounds did
not crystallize leading only to intensively red-colored solu-
tions. The molecular structure of the red compound 4a is
presented in Figure 9 (crystallographic data are reported in
the Supporting Information).
The oxidation of the alcoholate leads to a copper(I) spe-
cies, which may now form a copper(I) complex with the new
Schiff base ligand benzyl-(2-benzylimino-1-methyl-propyli-
dene)-amine and CuCl2 as anion. It is remarkable that the
compound 7, cluster II and the red material 4a ([Cu-
À
AHCTUNGTRENNG(UN benz2mpa)2]CuCl2 described above) always seem to occur
together in solution. This strongly indicates the existence of
a redox equilibrium between these species. Although in the
case of denatured ethanol we were able to identify the red
species (compound 4a) we cannot assign this structure to
the red species (compound 4) in other solvents due to lack-
ing of 2-butanone in solution. But it is most likely that simi-
lar redox reactions take place, which will lead to related
copper(I) complexes.
The strong redox activities of cluster I as well as of cluster
II were further demonstrated in a series of experiments:
1) It was observed that storing either of the clusters in de-
natured ethanol led to selective oxidation of some of the
solvent to acetaldehyde. Acetic acid was not detected.
2) Cluster I showed activity in oxidation of catechol and de-
rivatives such as 2-aminophenol. Adding a small amount
of cluster I to a solution of catechol in DMF under at-
mospheric conditions immediately led to the formation
of the red-colored corresponding quinone. UV/Vis-meas-
urements confirmed these observations (for spectra see
the Supporting Information). Investigations of the de-
tailed reaction mechanism are in progress.
3) Furthermore, cluster I can support hydroxylation of ali-
À
phatic C H bonds with dihydrogen peroxide as a co-re-
agent. Mixing a suspension of cluster I in acetonitrile
with cyclohexane or hexane and H2O2 will lead to the ac-
cording alcohols and side products (Scheme 1). During
this reaction overall yields for the oxidation of cyclohex-
ane up to 18% and a turnover of 72 could be reached.
For the oxidation of n-hexane overall yields of 6% and a
turnover of 24 could be observed (no regioselectivity was
Figure 9. Molecular structure of [CuACHTNUTRGNEUNG(benz2mpa)2]CuCl2 (compound 4a,
benz2mpa=benzyl-(2-benzylimino-1-methyl-propylidene)-amine), hydro-
gen atoms are omitted for clarity Ellipsoids are drawn at the 50% proba-
bility level. An additional Lewis structure of the ligand is presented for
clarity.
Scheme 1. Hydroxylation of cyclohexane and n-hexane by cluster I.
5346
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5342 – 5351

Reactions of Copper(II) Chloride in Solution
FULL PAPER
observed). In contrast to the oxidation of cyclohexane
the dihydroxylated product could also be found; which is
about 30% of the product. This behavior resembles a
multicopper system found by Kirillov et al.[4a] Similar to
these authors we could also observe the formation of an
activated copper(I) system by the addition of H2O2. As
proposed in the mechanism by Kirillov et al. we could
also observe a strong evolution of molecular oxygen, in-
dicating the reduction of the copper(II) species and
hence, formation of an activated copper(I) system. Addi-
tionally, the reaction mixture turned red. As suspected
we could confirm the formation of the copper(I) system
due to the strong reaction with cuproine, which is a
highly selective detection reagent for copper(I) ions. In-
vestigations of the detailed reaction mechanism and
studies on increasing yields are in progress. Preliminary
results are presented in Table 1.
Figure 10. Molecular structure of the salt benzyl-(1,1-dimethyl-3-oxo-
butyl)-ammonium chloride, HC13H19NOCl. Ellipsoids are drawn at the
50% probability level.
pounds or due to catalysis with strong bases such as KOH.
In our case neither acetone is very CH acidic nor is benzyl-
AHCTUNGTREGaNNNU mine a strong base. It is quite interesting to compare this
reaction with the reaction of cluster I with 2-buta-
none described above. The only difference between
2-butanone and acetone is a methyl group. Thus, it
is surprising to observe such a different behavior.
However, both reactions have in common that the
products may be isolated in good and reproducible
yields. Kinetic studies are in progress to obtain de-
tailed information on the reaction mechanisms,
however, these investigations are complicated and
time consuming (see below).
Table 1. Hydroxylation of cyclohexane with H2O2 and cluster I.
Catalyst
(amount
Starting
material
ACHTUNGTRENNUNG(amount ACHTUNGTRNE[NUGN mmol])
H2O2
[mmol]
Yield
alcohol
[%]
Yield
ketone
[%]
Turnover
N
[mmol])
ACHTUNGTRENNUNG
cluster I (25)
cluster I (25)
CuCl2·2H2O (41)
–
cyclohexane (10)
cyclohexane (10)
cyclohexane (10)
cyclohexane (10)
n-hexane (10)
10
20
20
20
10
7
14
7
0
6
3
4
4
0
0
40
72
27
–
cluster I (25)
24
4) When dihydrogen peroxide was added to a suspension of
cluster I in acetonitrile without substrate a violent reac-
tion occurred and a brown material, compound 9, could
be isolated (Figure 3). So far, this compound could not
be characterized. By using tert-butylhydroperoxide in-
stead of dihydrogen peroxide allowed us to selectively
oxidize p-chlorbenzylic alcohol catalytically to the ac-
cording aldehyde in excellent yields (yields up to 90%,
turnover of 360, for details, see the Supporting Informa-
tion) according to the Equation (3):
Scheme 2. Proposed a) aldol condensation and b) Michael addition for
the formation of HC13H19NOCl.
Furthermore, an analogous reaction behavior could be ob-
served by using zinc chloride instead of copper(II) chloride.
Colorless crystals could be isolated in good yields from the
solution and were structurally characterized. As observed
for the reaction with copper chloride zinc chloride also
seems to facilitate an aldol condensation and a following
Michael addition (see the Supporting Information for a de-
tailed description of this reaction).
It is important to point out that both CuCl2 as well as
ZnCl2 seem to allow reactions to proceed at room tempera-
ture that normally only may be achieved under much harsh-
er conditions. It is an interesting future goal to investigate if
zinc chloride can form cluster units analogous to Cu4OCl6L4
as well. The reactivity of these clusters should be different
with regard to the non-redox behavior of the zinc ion in
contrast to the observed CuI/CuII redox couple. This would
allow to separately investigate all reactions that do not in-
clude changes in the redox state of the metal ion.
Another quite interesting reaction was observed when
cluster I was dissolved in acetone. The solution turned to a
red color within a few hours and colorless crystals could be
separated from the mixture. The molecular structure of this
compound is shown in Figure 10 (crystallographic data are
reported in the Supporting Information).
Obviously the cluster facilitated an aldol condensation
(Scheme 2a) followed by
(Scheme 2b).
a Michael addition reaction
This reactivity is quite unusual, because aldol condensa-
tions are usually observed either with rather CH-acidic com-
Chem. Eur. J. 2013, 19, 5342 – 5351
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5347

S. Schindler et al.
During our investigations we went back to original litera-
ture reports on clusters of the type Cu4OX6L4. Although
most of these complexes have been studied thoroughly by
spectroscopy only few crystal structures were reported such
as [Cu4OX6(py)4] (py=pyridine).[1g]
At this point it is important to clarify that not all co-li-
gands seem to support formation of such cluster units. When
we used aniline instead of benzylamine the expected copper
clusters of this type the central oxygen atom is coordinated
tetrahedrally to four copper(II) ions, whereas the six chlo-
AHCTUNGTREGrNNNU ide ions form the typical cage for these structures. Every
copper ion is coordinated to three chloride anions, the cen-
tral oxygen atom and one ligand (in this case methanol),
which results in a trigonal-bipyramidal coordination for
each copper ion.
Similar to methanol other solvents can be used and we
successfully could structurally characterize Cu4OCl6L4 (L=
acetonitrile, DMSO, and DMF). The molecular structures of
these compounds are presented in the Supporting Informa-
tion.
complex [CuACHTUNGTRENNUNG(aniline)2Cl2] immediately formed and here we
could not observe cluster units confirming previous results
by tom Dieck et al.[1f] So far we do not have an explanation
for this observation. The molecular structure of [Cu-
A
We think that these “solvent” cluster complexes are quite
important for the general understanding of stoichiometric
are reported in the Supporting Information).
and catalytic reactions using copperACTHNUTRGNEUNG(I/II) chloride as an
active compound. Most likely there are no simple copper(II)
or copper(I) ions present in solution in the presence of
chloride ions (or bromide ions). Formation of different clus-
ter units will occur in the majority of these reactions (see
below). Depending on the reaction conditions copper chlo-
A
6ACHTNUGTREN(UNGN MeOH)4] cluster in meth-
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
obtain some of the other solvent clusters.[1c] As a result of
the molecular formula the presence of oxygen (e.g., in form
of H2O) is indispensable for the formation of these clusters.
To get an idea which complexes copper chloride can form in
methanol when an oxygen source is excluded, we carried
out an experiment where CuS (instead of CuO) and CuCl2
were heated to reflux in dry methanol (experimental condi-
Figure 11. Molecular structure of the compound [CuACHTNURGTENUNG(aniline)2Cl2]. Hy-
drogen atoms omitted for clarity. Ellipsoids are drawn at the 50% proba-
bility level.
However, most interestingly to us seemed the cluster
[Cu4OX6ACHTUNGTRENNUNG(MeOH)4] in which the solvent molecules form the
additional ligands in the cluster unit. We could easily pre-
pare this cluster under inert conditions according to the pro-
cedure published previously by tom Dieck et al. and ob-
tained crystals suitable for crystallographic characteriza-
tion.[1c] The molecular structure presented in Figure 12 (crys-
tallographic data are reported in the Supporting Informa-
tion) confirmed the original postulation including the two
solvent molecules, which co-crystallized with the cluster.
The yellow crystals are sensitive towards air and moisture
and therefore have to be kept under argon. As for the other
tions analogous to the synthesis of [Cu4OCl6ACTHNGUETRNU(NG MeOH)4]).
Crystals could be isolated in very good yields from the solu-
tion and the molecular structure of this compound is pre-
sented in Figure 13 (crystallographic data are reported in
the Supporting Information).
Figure 13. Molecular structure of the complex [Cu
soids are drawn at the 50% probability level.
ACHTUNGRTEN(UNGN MeOH)2Cl2]. Ellip-
The structure shows a square-planar-coordinated cop-
AHCTUNGTRENNUNG
the structure motif [CuACHTUNRGTNE(UNG H2O)2Cl2] is most common and was
found quite often to co-crystallize with other compounds.[11]
However, so far it was not possible to crystallize [Cu-
AHCTUNGTRENNUNG
Figure 12. Crystal structure of the compound Cu4OCl6L4 (L=methanol).
Solvent molecules are omitted for clarity. Ellipsoids are drawn at the
50% probability level.
AHCTUNGTRENNUNG
5348
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5342 – 5351

Reactions of Copper(II) Chloride in Solution
FULL PAPER
trile, DMF, acetone or THF of the formula [Cu(L2)Cl2] (L=
solvent molecule) are not known yet.
Although not discussed in detail, several of our findings
should be relevant to reactions supported by CuCl as
well.[15b,16] The redox reactions as described above lead to
formation of CuI from CuII and vice versa. In some publica-
tions it is mentioned that either CuCl or CuCl2 can be
used.[15a]
It is important to note that an “outstanding” copper cata-
lyst should always be compared to simply adding copper(II)
chloride (or copper(I) chloride). As discussed above this
might show the same reactivity due to the formation of
active species in solution (sometimes most likely leading to
the same intermediate that is formed with the catalyst). In
contrast, copper(II) salts such as sulphate or triflate in the
absence of chloride ions may not be very effective for many
catalytic reactions.
Furthermore, it should be pointed out that the complex
behavior described above for our copper benzylamine clus-
ter is not restricted to this system but will occur with other
metal ions and ligands as well. Although we cannot say that
we have completely understood these reactions yet, we have
provided new insight into the formation and reactivity of a
CuCl2/ligand system. It is clear that a better and more de-
tailed understanding of these reactions will allow the devel-
opment of optimized and new copper catalytic systems.
We always tested the catalytic activity of cluster I in com-
parison by simply adding CuCl2·2H2O alone in all of our ex-
periments. Similar to previous results we observed that
CuCl2·2H2O is catalytically active in most of these reactions
as well (only some selected examples are given in the refer-
ences).[13] However, this is not surprising because as dis-
cussed above there is no simple CuCl2 in solution. Different
clusters can be formed and even the solvent can act as a
ligand. Thus, CuCl2 can form Cu4OX6L4 with L=solvent.
Here, the solvent can be easily replaced by other stronger
coordinating ligands, such as amines or chloride ions.[1c]
Despite the fact that we could synthesize cluster I—or
that in general cluster units of the type Cu4OX6L4 (X=halo-
gen, L=ligand or X) can be isolated—it is still questionable
if these species are really the active catalysts. Excellent de-
tailed and complicated kinetic studies by Davis and co-
workers in the past have demonstrated that in solution quite
a large number of equilibria between different cluster units
(or cluster parts) can be observed (as one example the
chloride only cluster Cu4OCl10 should be mentioned) that
actually may become the reactive species.[14] To really under-
stand these reactions in detail will need a lot of more re-
search efforts in that area. However, we think with combin-
ing the previous excellent results of different research
groups and our new findings that we moved one step closer
to a better understanding of these reactions.
Experimental Section
Analytical data (elemental analyses, IR spectra, and crystallographic
data[17]) are summarized in the Supporting Information.
Preparation of cluster I: CuCl2·2H2O (15.3 g, 90 mmol) was dissolved in
methanol (ca. 50 mL) and benzylamine (10.7 g, 100 mmol) was added
dropwise to the solution under stirring. The mixture was stirred for ap-
proximately 10 min and the resulting olive green precipitate was filtered
off and dried to give cluster I (15.5 g, 7.13 mmol, 71%). The preparation
of the bromide analogue of cluster I was performed accordingly except
that CuBr2 was used instead of CuCl2·2H2O. The target compound was
obtained in 80% yield.
Conclusion
We have obtained a useful catalytic system from copper(II)
chloride and benzylamine. By using cluster units derived
from this system we were able to perform a series of inter-
esting reactions, that is:
Preparation of cluster II: 2-Butanone (0.72 g, 10 mmol) was dissolved in
ethanol (100 mL) and benzylamine (1.1 g, 10 mmol) was added dropwise
to the stirred solution. Afterwards, CuCl2·2H2O (0.85 g, 5.0 mmol) was
added. The solution was stirred for a few minutes under air and then al-
lowed to stand until crystallization occurred. The flask was opened every
day and shaken slightly. After a few days the solution turned to a red
color and green crystals could be obtained in good yields from the solu-
tion.
1) selective oxidation of ethanol
2) oxidation of catechol and derivates
À
3) hydroxylation of aliphatic C H bonds with H2O2
4) selective oxidation of p-chlorbenzylic alcohol with tert-
butylhydroperoxide.
In addition we could show that the cluster units are highly
reactive and show a huge affinity towards carbonyl com-
pounds such as 2-butanone or acetone. We could observe
different types of reactions starting with the formation of a
Schiff base. As consecutive reactions we could observe se-
lective hydroxylation of aliphatic C H bonds as well as an
aldol condensation and Michael addition under mild condi-
tions.
It is clear from our work and literature reports that the
behavior of simple copper(II) compounds in solution can be
quite complex. Many times our understanding of CuCl2 in
solution seems to be oversimplified (organic chemists tend
to write CuCl2 in chemical equations just above the reaction
arrow; only some examples are given in the references).[13,15]
Preparation of compound 1: CuCl2·2H2O (0.85 g, 5.0 mmol) was dis-
solved in methanol (ca. 20 mL) and benzylamine (2.1 g, 20 mmol) was
added dropwise to the stirred solution. The solution was stirred for ap-
proximately 10 min and the precipitate was filtered off. The filtrate was
added to diethyl ether (ca. 40 mL) to precipitate the rest of compound 1
(1.4 g, 4.0 mmol, 80%).
À
Preparation of compound 2: CuCl2·2H2O (0.51 g, 3.0 mmol) was dis-
solved in methanol (ca. 15 mL) and benzylamine (3.2 g, 30 mmol) was
added dropwise to the stirred solution. The resulting blue solution was
added to diethyl ether (ca. 50 mL) and allowed to stand until crystalliza-
tion occurred. The crystals were isolated and washed with little portions
of diethyl ether to give compound 2 (1.4 g, 2.4 mmol, 81%).
Preparation of compound 3: CuCl2·2H2O (3.4 g, 20 mmol) was dissolved
in methanol (ca. 20 mL) and benzylamine (0.54 g, 5.0 mmol) was added
dropwise to the stirred solution. The solution was stirred for 5 min and
then added to diethyl ether (ca. 40 mL) to precipitate compound 3. The
Chem. Eur. J. 2013, 19, 5342 – 5351
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5349

S. Schindler et al.
precipitate was isolated and washed with small portions of diethyl ether
to give compound 3 (0.93 g, 2.20 mmol, 88%).
Preparation of methanol cluster: The synthesis was performed according
to literature:[1c] CuCl2 (54 g, 0.4 mol) and CuO (11 g, 0.14 mol) were sus-
pended in dry methanol (100 mL) and heated to reflux under inert condi-
tions for 2 h. Afterwards the hot reaction mixture was filtrated and the
filtrate was left in the fridge for crystallization. The crystals were isolated
and dried in vacuum to remove the crystal methanol to give the target
cluster (61 g, 0.1 mol, 74%).
Preparation of compound 4: Cluster I was prepared as described above
in ethanol but this time it was not isolated. The solution turned to a red
color after two days. Then the precipitate was filtered off and the red-col-
ored filtrate was evaporated in a rotary evaporator. The resulting red-col-
ored oil was dissolved in ethanol (1–2 mL) and added to diethyl ether
(ca. 350 mL). The red-colored precipitate was filtered off. Compound 4 is
slightly sensitive towards air and therefore was stored under argon. It is
possible to use methanol or acetonitrile as well to prepare a red-colored
solid as described.
Preparation of acetonitrile, DMSO, and DMF cluster: According to the
literature[1c] the methanol cluster (2.5 g) was dissolved in either acetoni-
trile, DMSO, or DMF (5 mL) and heated to reflux under inert conditions
for 30 min. The solvent was then removed in vacuum and the pure prod-
ucts were achieved in 95% yield.
Preparation of compound 5: As described above cluster I was prepared
in ethanol but again not isolated. After the solution turned to a red
color, air was bubbled through the suspension for approximately 30 min
until the solution turned to an orange color. Similar to the preparation of
compound 4 the precipitate was now filtered off and the filtrate was
evaporated in a rotary evaporator. The orange-colored oil obtained was
dissolved in ethanol (1–2 mL) and added to diethyl ether (ca. 350 mL).
The orange-brown-colored precipitate was then isolated. Compound 5 is
slightly sensitive towards air and therefore was stored under argon.
Preparation of [CuACHTUNRTGNE(UNG MeOH)2Cl2]: CuS (6.7 g, 0.07 mmol) and CuCl2 (27 g,
0.2 mmol) were suspended in dry methanol (50 mL) and heated to reflux
under inert conditions for 2 h. Afterwards the hot reaction mixture was
filtrated and the filtrate was left in the fridge for crystallization to give
[Cu
ACHTUNGTRENNUNG
Preparation of [CuAHCTUNGTRENNUNG
solved in methanol (ca. 20 mL) and aniline (0.52 g, 5.6 mmol) dissolved
in a small portion of methanol was added dropwise to the stirred solu-
tion. The suspension was stirred for approximately 5 min and the precipi-
tate was then isolated and washed with small portions of methanol. The
product was dried in air (0.74 g, 2.3 mmol, 46%).
Preparation of compound 6: Preparation was performed similar to the
preparation of cluster I except that 20% of the solvent was replaced by
water.
Preparation of compound 7: 2-Butanone (0.72 g, 10 mmol) was dissolved
in ethanol (100 mL) and benzylamine (1.1 g, 10 mmol) was added drop-
wise to the stirred solution. Afterwards CuCl2·2H2O (0.85 g, 5.0 mmol)
was added. The obtained suspension was stirred for a few minutes in a
closed flask. The flask was not opened for a few days. During this time
the solution has turned to a red color and the olive-colored precipitate
turned to colorless crystals. The resulting crystals were sensitive towards
air and therefore were stored under argon.
Acknowledgements
S. Lçw and S. Schindler are thankful to the Stiftung Stipendienfonds der
Chemischen Industrie e.V. for providing a Ph.D. scholarship to S. Lçw.
Furthermore, we appreciate private communications from Prof. Dr. Hein-
dirk tom Dieck and Dr. Hans-Peter Brehm.
Preparation of compound 8: A small portion of cluster I was suspended
in methanol and solid NaBH4 was slowly added to the suspension until a
colorless solution and black precipitate formed. The solid was isolated
and dried in air.
[1] a) J. A. Bertrand, J. A. Kelley, J. Am. Chem. Soc. 1966, 88, 4746–
4747; b) J. A. Bertrand, J. A. Kelley, Inorg. Chem. 1969, 8, 1982–
1985; c) H. tom Dieck, H.-P. Brehm, Chem. Ber. 1969, 102, 3577–
3583; d) J. A. Bertrand, J. A. Kelley, Inorg. Chim. Acta 1970, 4, 526–
528; e) H. tom Dieck, Inorg. Chim. Acta 1973, 7, 397–403; f) H.
Bock, H. tom Dieck, H. Pyttlik, M. Schnçller, Z. Anorg. Allg.
Chem. 1968, 357, 54–61; g) B. T. Kilbourn, J. D. Dunitz, Inorg.
Chim. Acta 1967, 1, 209–216; h) J. A. Bertrand, Inorg. Chem. 1967,
6, 495–498; i) J. J. De Boer, D. Bright, J. N. Helle, Acta Crystallogr.
Sect. B 1972, 28, 3436–3437; j) N. S. Gill, M. Sterns, Inorg. Chem.
1970, 9, 1619–1625.
[2] a) F. A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, 5th
ed., Wiley, New York, 1988; b) J. A. Bertrand, J. A. Kelley, C. E.
Kirkwood, J. Chem. Soc. Chem. Commun. 1968, 1329–1330; c) M.
Dey, C. P. Rao, P. K. Saarenketo, K. Rissanen, Inorg. Chem.
Commun. 2002, 5, 380–383; d) C. Mukherjee, T. Weyhermꢀller, E.
Bothe, E. Rentschler, P. Chaudhuri, Inorg. Chem. 2007, 46, 9895–
9905; e) S. Thakurta, P. Roy, R. J. Butcher, M. S. El Fallah, J. Ter-
cero, E. Garribba, S. Mitra, Eur. J. Inorg. Chem. 2009, 4385–4395;
f) R. Wegner, M. Gottschaldt, H. Gçrls, E.-G. Jꢁger, D. Klemm,
Chem. Eur. J. 2001, 7, 2143–2157.
Preparation of compound 9: Compound 9 was prepared in situ by reduc-
ing a mixture of cluster I in acetonitrile with a 30% aqueous solution of
H2O2. The brown-red-colored precipitate was filtered off. Compound 9 is
highly sensitive towards air and unstable at room temperature.
Oxidation of ethanol: A small portion of cluster I (or cluster II) was sus-
pended in denatured ethanol (ca. 5–10 mL) and allowed to stand for a
day. Acetaldehyde was detected by GC-MS and exact mass detection.
Oxidation of catechol derivatives: A small amount of o-aminophenol (or
catechol) was dissolved in DMF and a tip of a spatula of cluster I was
added. The solution immediately turned to a red color and was investi-
gated by UV/Vis measurements.
Hydroxylation of cyclohexane and n-hexane: Cluster I (54 mg, 25 mmol)
was suspended in acetonitrile (10 mL) and cyclohexane (0.84 g, 10 mmol)
was added. A dihydrogen peroxide solution (30%, 2 mL, ca. 20 mmol
H2O2) was added dropwise to the stirred solution. A reflux condenser
was attached to the flask to prevent evaporation of cyclohexane due to
the reaction heat. The solution was stirred until gas evolution stopped
and was investigated by GC-MS. Hydroxylation of n-hexane was investi-
gated in the same way.
[3] a) M. E. Lines, A. P. Ginsberg, R. L. Martin, R. C. Sherwood, J.
Chem. Phys. 1972, 57, 1–19; b) H. Wong, H. tom Dieck, C. J.
OꢄConnor, E. Sinn, J. Chem. Soc. Dalton 1980, 786–789; c) W. J.
Jones, S. Gupta, L. J. Theriot, F. T. Helm, W. A. Baker, Inorg. Chem.
1978, 17, 87–90; d) J. Reim, K. Griesar, W. Haase, B. Krebs, Dalton
Trans. 1995, 2649–2656; e) H. Oshio, Y. Saito, T. Ito, Angew. Chem.
1997, 109, 2789–2791; Angew. Chem. Int. Ed. Engl. 1997, 36, 2673–
2675; f) J. W. Hall, W. E. Estes, E. D. Estes, R. P. Scaringe, W. E.
Hatfield, Inorg. Chem. 1977, 16, 1572–1574; g) J. Barnes, G. W. In-
man, Jr., W. E. Hatfield, Inorg. Chem. 1971, 10, 2355.
Oxidation of p-chlorbenzylic alcohol: Cluster I (54 mg, 25 mmol) was sus-
pended in dichloromethane (10 mL) and p-chlorbenzylic alcohol (1.4 g,
10 mmol) was added. To the stirred solution either tert-butylhydroperox-
ide (28,5%, 3 mL 10 mmol) dissolved in dichloromethane or tert-butylhy-
droperoxide (70%, 1.5 mL, ca. 10 mmol) dissolved in water was added.
The solution was stirred for a defined time and then the cluster was re-
moved through a chromatography column and the yield of the corre-
sponding aldehyde was determined by GC-MS. The results are presented
in the Supporting Information.
Preparation of C13H20NOCl: A saturated solution of cluster I in acetone
was prepared and left in the fridge for crystallization. The solution dyed
red after 1–2 h and colorless crystals were obtained in good yields.
[4] a) M. V. Kirillova, Y. N. Kozlov, L. S. Shul’pina, O. Y. Lyakin, A. M.
Kirillov, E. P. Talsi, A. J. L. Pombeiro, G. B. Shul’pin, J. Catal. 2009,
268, 26–38; b) A. M. Kirillov, M. N. Kopylovich, M. V. Kirillova,
5350
www.chemeurj.org
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 5342 – 5351

Reactions of Copper(II) Chloride in Solution
FULL PAPER
E. Y. Karabach, M. Haukka, M. F. C. Guedes da Silva, A. J. L. Pom-
beiro, Adv. Synth. Catal. 2006, 348, 159–174; c) S. Jammi, T. Pun-
niyamurthy, Eur. J. Inorg. Chem. 2009, 2508–2511; d) H. Sun, K.
Harms, J. Sundermeyer, J. Am. Chem. Soc. 2004, 126, 9550–9551;
e) A. M. Kirillov, M. N. Kopylovich, M. V. Kirillova, M. Haukka,
M. F. C. Guedes da Silva, A. J. L. Pombeiro, Angew. Chem. 2005,
117, 4419–4423; Angew. Chem. Int. Ed. 2005, 44, 4345–4349.
[5] K. D. Karlin, S. Itoh, Copper–Oxygen Chemistry, 4th ed., Wiley,
New Jersey, 2011, pp. 131–168
[6] a) C. Wꢀrtele, O. Sander, V. Lutz, T. Waitz, F. Tuczek, S. Schindler,
J. Am. Chem. Soc. 2009, 131, 7544–7545; b) C. Wꢀrtele, E.
Gaoutchenova, K. Harms, M. C. Holthausen, J. Sundermeyer, S.
Schindler, Angew. Chem. 2006, 118, 3951–3954; Angew. Chem. Int.
Ed. 2006, 45, 3867–3869; c) M. Weitzer, S. Schindler, G. Brehm, S.
Schneider, E. Hçrmann, B. Jung, S. Kaderli, A. D. Zuberbꢀhler,
Inorg. Chem. 2003, 42, 1800–1806; d) M. Schatz, M. Becker, O.
Walter, G. Liehr, S. Schindler, Inorg. Chim. Acta 2001, 324, 173–
179; e) S. Schindler, Eur. J. Inorg. Chem. 2000, 2311–2326.
[7] G. Zi, L. Xiang, Y. Zhang, Q. Wang, Z. Zhang, Appl. Organomet.
Chem. 2007, 21, 177–182.
Watson, Acta Crystallogr. Sect. B 1971, 27, 1619–1624; c) J. Giantsi-
dis, M. M. Turnbull, C. P. Landee, F. M. Woodward, J. Coord. Chem.
ˇ
ˇ ˇ
ˇ
2002, 55, 393–402; d) M. Dunaj-Jurco, J. Kozꢅsek, A. Ancyskina, G.
ˇ
ˇ
ˇ
Ondrejovic, D. Makꢆnovꢆ, J. Gazo, J. Chem. Soc. Chem. Commun.
1979, 654–655.
[12] R. D. Willett, K. Chang, Inorg. Chim. Acta 1970, 4, 447–451.
[13] a) A. Laouiti, M. M. Rammah, M. B. Rammah, J. Marrot, F. Couty,
G. Evano, Org. Lett. 2012, 14, 6–9; b) W. Wu, X. He, Z. Fu, Y. Liu,
Y. Wang, X. Gong, X. Deng, H. Wu, Y. Zou, N. Yu, D. Yin, J. Catal.
2012, 286, 6–12; c) Y. Ito, T. Konoike, T. Harada, T. Saegusa, J. Am.
Chem. Soc. 1977, 99, 1487–1493; d) H. Sun, X. Li, J. Sundermeyer,
J. Mol. Catal. A 2005, 240, 119–122; e) S. Kim, P. H. Lee, J. Org.
Chem. 2012, 77, 215–220; f) Y. Liu, J.-W. Sun, J. Org. Chem. 2012,
77, 1191–1197.
[14] a) A. El-Toukhy, G.-Z. Cai, G. Davies, T. R. Gilbert, K. D. Onan, M.
Veidis, J. Am. Chem. Soc. 1984, 106, 4596–4605; b) M. El-Sayed, A.
Abu-Raqabah, G. Davies, A. El-Toukhy, Inorg. Chem. 1989, 28,
1909–1914, and references therein.
[15] a) Q. Liu, P. Wu, Y. Yang, Z. Zeng, J. Liu, H. Yi, A. Lei, Angew.
Chem. 2012, 124, 4744–4748; Angew. Chem. Int. Ed. 2012, 51, 4666–
4670; b) M. C. Kozlowski in Copper Oxygen Chemistry (Eds.: K. D.
Karlin, S. Itoh), Wiley, New York, 2011, pp. 361–444; c) E. Boess, C.
Schmitz, M. Klussmann, J. Am. Chem. Soc. 2012, 134, 5317–5325;
d) H.-Q. Do, H. Tran-Vu, O. Daugulis, Organometallics 2012, 31,
7816–7818.
[16] a) M. Largeron, M.-B. Fleury, Angew. Chem. 2012, 124, 5505–5508;
Angew. Chem. Int. Ed. 2012, 51, 5409–5412; b) C. P. Frazier, J. R.
Engelking, J. R. deAlaniz, J. Am. Chem. Soc. 2011, 133, 10430–
10433.
[8] R. Belford, D. E. Fenton, M. R. Truter, J. C. S. Dalton 1972, 2345–
2350
[9] a) H. Remy, G. Laves, Ber. Dtsch. Chem. Ges. 1933, 401–407;
b) R. D. Willett, M. L. Larsen, Inorg. Chim. Acta 1971, 5, 175–179.
[10] a) S. H. Oakley, M. P. Coles, P. B. Hitchcock, Inorg. Chem. 2004, 43,
5168–5172; b) S.-L. Zheng, M. Messerschmidt, P. Coppens, Angew.
Chem. 2005, 117, 4690–4693; Angew. Chem. Int. Ed. 2005, 44, 4614–
4617; c) M. Angels Carvajal, S. Alvarez, J. J. Novoa, Chem. Eur. J.
2004, 10, 2117–2132; d) G. M. Chiarella, D. Y. Melgarejo, A. Rozan-
ski, P. Hempte, L. M. Perez, C. Reber, J. P. Fackler, Jr., Chem.
Commun. 2010, 46, 136–138.
[17] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[11] a) M. Bukowska-Strzyzewska, J. Skoweranda, Acta Crystallogr. Sect.
Received: October 29, 2013
C
1987, 43, 2290–2293; b) R. J. Williams, D. T. Cromer, W. H.
Published online: February 28, 2013
Chem. Eur. J. 2013, 19, 5342 – 5351
ꢂ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
5351